Studies are proposed in monkeys, human controls and patients which explore mechanisms whereby the cerebellum may both coordinate and adapt movement. The cerebellum works through downstream movement generators in motor cortex, red, reticular and vestibular nuclei and spinal cord to control movement. What it adds to their functional capacity is debated. Hypotheses to be tested are that the cerebellum can : 1) combine the actions of small synergic units into larger units; 2) modulate downstream generators to make their output specifically appropriate for each of various tasks, and 3) adaptively create new synergic units and modify preexisting ones in order to change movement performance according to change in task requirement. One project tests whether cerebellar neurons fire during and inactivation impairs a monkey's pinch (thumb and forefinger in combination) preferentially to thumb or forefinger flexion alone. The premise is that motor cortex can alone control single digit movements, and that cerebellum may be required to add them and thus create the synergy. A second project tests whether cerebellar neurons fire during and inactivation impairs the monkey's coordinated reach preferentially to the component movements of turning of eye and head and movements of shoulder, elbow, and wrist. The premise is that brainstem mechanisms alone can generate reach (e.g., tonic neck reflex), and that cerebellum may be required to adjust the synergy, without actually creating it. A third project tests whether cerebellar neurons fire during and inactivation impairs preferentially the monkey's adjustment of the alignment of eye and hand in pointing to a visual target preferentially to the performance of this act. Previous results suggest that the cerebellum helps adapt this behavior; it is unclear to what extent it controls the adjusted performance. AU three projects address the question of location of the mechanisms (cortex vs. nuclei), and what cellular elements are involved. Human normals and cerebellar patients perform similar tasks to validate monkey results. These data will be used both to test a scientific model of the mechanisms of cerebellar control and a clinical model for the training of cerebellar patients disabled by incoordination to use compensatory single-jointed movement strategies in their rehabilitation toward independent living.